The impact of the GDP and population on trade of COMESA using panel data approach

Abstract

The general consensus from a plethora of literature is that the estimates of variability for analyzing panel data models are precise and achieve accurate inferences. Panel data can also be used to control for heterogeneity through several models, such as the fixed effect model (FEM), the random effects model (REM), and the pooled regression model(PM). This study aims to uncover the increases in trade volume among the member countries of Common Market for Eastern and Southern Africa (COMESA) and identify the important factors that affect COMESA trade using panel data models. In addition, this study intends to investigate the independent variables and their effect on the export and import of COMESA member countries. Results show that in FEM, an increase in GDP leads to a 65% increase in imports, and in REM, an increase in GDP leads to an 86% increase in exports. These findings suggest that FEM and REM are appropriate for explaining the imports and exports of COMESA, respectively.